1. Science
  2. Publications
  3. Information Processing Systems
  4. 4(78)'2009
  5. Finding minimal sets of features for describing discrete information objects in database

Finding minimal sets of features for describing discrete information objects in database

D.E. Sitnikov, O.A. Romanenko, E.V. Titova
Annotations languages:

Finding of dependences in information as logical rules is one of typical tasks of the Intellectual analysis of data (Data Mining), at the decision of which very actual is limitation of amount of signs, included in a resulting rule. Determination of importance of one or another sign allows to shorten the amount of the analysed information and as a result, to shorten the amount of logical terms in the found rules.
Keywords: logical rules, power of approximation, meaningfulness of sign, fuzzy sets, overhead and lower approximations, importance of sign